> For the complete documentation index, see [llms.txt](https://academy.shade.inc/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://academy.shade.inc/ai-tools/the-shade-mcp.md).

# The Shade MCP

The Shade MCP connects your workspace directly to the AI tools and agents you already use , Claude, ChatGPT, Cursor, or any other MCP compatible client. Once connected, your assistant can search assets, read and update custom objects, build automations, and manage sharing and metadata, all from a prompt, without opening Shade.

> Note: The Shade MCP is different from the Academy Shade MCP, which only answers questions about Shade's documentation. The Shade MCP connects to your live workspace and can read and modify your actual data.

{% embed url="<https://youtu.be/lAas9Sekjvk?si=n-h5UwLaK1VlYi_R>" %}

### Getting Started

Connect the Shade MCP from **Settings → Integrate → MCP connections**. You'll see pre-built options for the platforms we see used most often. If your platform isn't listed, copy the general MCP URL from this page and paste it into your client's custom connector field.

<figure><img src="https://450378260-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FUNq6h2Bfw7aAK0J2ffnZ%2Fuploads%2Ff78z6qfRtR9BggQRugEM%2FScreenshot%202026-07-14%20at%205.15.39%E2%80%AFPM.png?alt=media&amp;token=8bbbd24e-8a35-49c4-9c55-5f7b04f7d0ed" alt=""><figcaption></figcaption></figure>

#### Claude

1. In Shade, go to **Settings → Integrate → MCP connections**.
2. Select **Claude**.
3. Follow the prompt to open Claude's connector settings.
4. Paste the provided URL into the custom connector field.
5. Sign in through OAuth to authorize the connection.
6. Test it with a prompt like:

   <mark style="color:red;">`Find all assets tagged "hero shot" in my Fall Campaign drive`</mark>

   <mark style="color:red;">`Create a share link for this folder with comment permissions`</mark>

#### ChatGPT

1. In Shade, go to **Settings → Integrate → MCP connections**.
2. Select **ChatGPT**.
3. Open your **Tools** or **Connectors** settings in ChatGPT.
4. Paste the provided URL when adding a custom connector.
5. Authorize the connection.
6. Test it with a prompt like:

   <mark style="color:red;">`Search my workspace for last week's uploads`</mark>

   <mark style="color:red;">`List the custom objects in my Clients drive`</mark>

#### Cursor

1. In Shade, go to **Settings → Integrate → MCP connections**.
2. Select **Cursor**.
3. Open Cursor's **MCP** settings and add the server.
4. Paste the provided URL.
5. Authorize the connection.
6. Test it with a prompt like:

   <mark style="color:red;background-color:red;">`Pull the metadata schema for my Deliverables object`</mark>

#### Other MCP Clients

1. In Shade, go to **Settings → Integrate → MCP connections**.
2. Copy the general MCP URL.
3. Paste it into your client's custom connector or MCP server field.
4. Authorize the connection through OAuth.

<figure><img src="https://450378260-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FUNq6h2Bfw7aAK0J2ffnZ%2Fuploads%2Fo7Shys1WhrSIaQTkZQW2%2FScreenshot%202026-07-14%20at%205.18.27%E2%80%AFPM.png?alt=media&amp;token=dfa0fa53-d4c4-4bf6-9d46-b9f27b765232" alt=""><figcaption></figcaption></figure>

### What the Shade MCP Can Do

Once connected, you can review exactly what the MCP is allowed to do and set which actions require your approval from within your AI client's tool settings.

**Manage** (create, edit, and delete):

* Automations
* Collections
* Comments
* Files
* Metadata
* Custom objects
* Share Links
* Views

**Read,only:**

* Browse Shade
* Find duplicates
* Get assets and asset previews
* Get download URLs
* Search
* Access and read transcripts
* List workspaces and drives

### Example Workflows

**Auto sync project management tools with custom objects:** Connect another platform's MCP (Notion, Asana, ClickUp, HubSpot) alongside Shade's, then ask your LLM to pull the data from that tool and build a matching custom object in Shade. For example, "pull all the data from my Notion Content Calendar and set it up as a custom object, and sync both platforms daily at 5pm." Metadata fields and options carry over, and edits made on either platform reconcile on the schedule you set.

**Build an automation from a prompt:** Instead of building an automation block by block, describe the workflow directly. For example, "any time a file is uploaded to my rough exports folder, generate a share link with comment permissions and email it to <timothy@shade.inc> with the subject 'New Export Delivery.'" The MCP builds the automation to match.

**Move data between connected tools:** With Shade connected alongside other MCPs, an agent can act across platforms in one step, pulling a brief from a project management tool and gathering matching assets into a Shade collection, or pushing a generated share link back into an external tracker.

[Shade How to use the Shade MCP](https://www.youtube.com/playlist?list=PLWwQgtasVlKA)

### Things to Keep in Mind

The Shade MCP is separate from the **Academy Shade MCP**. Use the Academy MCP for documentation questions, and the Shade MCP for anything that touches your actual workspace.

### Related

[Custom objects](/automations-and-custom-objects/custom-objects.md)

[Automations](/automations-and-custom-objects/automations.md)

[Zapier Integration](/workspaces-and-drives/zapier-integration.md)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://academy.shade.inc/ai-tools/the-shade-mcp.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.